Two-tier research?

There’s always mutterings about forthcoming changes to REF, but speculation was particularly feverish in the lead up to this year’s Universities UK conference.

All kinds of people were confidently predicting a reversion to the rules of the previous exercise (James still speculates a version of this is a possibility), or at the very least a reweighting in favour of outputs and away from culture and environment. So the announcement, by Patrick Vallance, of a three-month pause (that would have no effect on the date of REF 2029) came as something of an anti-climax.

But Research England has just published details of what will be happening during this pause, and it’s all starting to feel a lot more interesting.

There’s still a lot to finalise about the way the next REF will run, and there’s clearly stuff to learn from the recent people, culture, and environment pilot. We can add to this, at a top level, a renewed interest in simplification and streamlining.

But there are also a number of work packages in the immediate future of Research England that feel rather more radical. And the further reforms that have been mooted feel sensible but the obvious question is why not do them in the first place.

The funding body will explore the option of baseline performance in research culture being a condition of funding. Some of the controversy about PCE has been about the way in which it can be assessed – how can you say with confidence that the culture at one provider is “better” than the culture at another? Academic judgement (the same by which panels assess outputs and impact case studies) is apparently not good enough in this situation – the argument goes that PCE is just a “tick-box” exercise.

The direction of travel appears to be that if you can’t tick enough of those boxes – which relate, in case we have forgotten, to the basics of running a department in a manner that actually supports researchers – you may get no funding at all. Quite how this plays out in the current DEI obsessed political environment will be interesting to watch. You can expect a lot of interest in precisely what constitutes a minimum viable research culture, and how compliance can be evidenced. It is also not clear that measuring a minimum baseline would be any easier, or any more convincing, than measuring culture within REF.

The other elephant in this particular indoor menagerie is whether adding a baseline compliance level means that assessment of practice above this benchmark needs to be in the main REF at all (or, less radically, to the same extent). If a part of the job of the REF is to foster healthy and sustainable research culture, then it is possible that a yes/no eligibility criteria does this just as well as a more detailed assessment. You might lose, of course, a chance to really drill in to the evidence – but it might be the kind of successful compromise that gets everyone (at least, everyone sensible) a little bit of what they want.

REF is applied in the same way (pretty much) to all universities and research performing organisations. A large research-intensive provider, with thousands of academics and research support staff, is much more able to churn out the documentation and evidence that REF needs than a place that – though it may still be doing world-class research – is comparatively under-staffed and under-resourced.

This is not a new problem, and there are probably no new solutions, but the news that Research England is exploring options for a twin- or multi-track lighter touch approach to assessment for less research-intensive institutions and/or smaller specialist institutions is likely to see ideas discussed that are less concerned with levelling the playing field and more concerned with building alternative playing fields. The arguments will come around the thresholds between the various tracks, and the validity (or otherwise) of feeding results in to funding allocations in the same way from different processes.

Likewise, three months to consider how funding allocation mechanisms in England could be modified to reward collaboration and specialisation (as part of the ongoing review of Strategic Institutional Research Funding) could mean a move away from the prevailing reverse Matthew principle approach of giving more to those that have a lot. It is probably right that a focus on the kind of cross-institutional, needs-focused research ties in neatly with government priorities – but surely research grants and projects (rather than QR) is the place to do this?

Accelerating current work and thinking about the future of research assessment feels like an invitation for more radicalism, and at greater speed, in the near future. If you are a research manager hoping for certainty and stability you may have to wait a little longer.